Data governance has become an overused term in our industry. Depending on context, the main responsibility of the data governance function can refer to any or all of:

  • Setting the direction of master data management efforts;
  • Deciding who in the organization will create, approve, enforce and monitor data; and
  • Determining who and what is involved in the workflow of data.

For the purposes of this column, I am going to call "governance" the body that sets the program direction (i.e., subject areas, sources, targets, business rules, etc.). Under this definition, the data governance function will necessarily permeate major areas of an organization to determine the who, what, when and where of information. Often the outcome of these questions will be implemented using the MDM workflow function. The participants in the workflow will be the data stewards. In this column, I am targeting the deciding or decision rights function when I refer to data governance.
Organizations need governance for many reasons. Often, the journey is intertwined with MDM. While you can have one without the other, the two disciplines are inextricably tied together. Data governance without MDM lacks a means of effective implementation. MDM without data governance is a data integration project using a tool in the MDM category. An effective project uses MDM as the execution arm for data governance decisions, but data governance can only support decision-making when the organization is ready to embrace the discipline required.

Data governance readiness sounds like one of those black hole, consultancy-created services in search of a business problem. But truly, without a gauge on organizational readiness, data governance efforts will misfire. Believing the organization is more ready for data governance than it is will lead to a governance model that looks great on paper but has no chance of success in the real world. Most organizations fall into this state of unreadiness. I have not yet met an organization that was ready to tackle governance without first making a concerted effort to prepare.

Consider the sheer intrusiveness of imposing structure upon the creation, deletion, approval and distribution rights of corporate information. It's quickly obvious that simply automating what is in place today is hardly adequate. Our current policies may not translate to the logical form necessary for automation, and translating poor practices to business rules will only raise concerns about current processes that need changing. Automation may be a step in a better direction, but there is no avoiding the fact that we need executive guidance to revisit our working processes if we hope to create real benefits from governance.

Case in point, client A has a problem with product data management. New product introduction cycles from their vendors are unnecessarily long, error-prone and dramatic. The marketing team's involvement in the process is unpredictable or opportunistic. The team often finds after the fact that it should have been involved and should have asked the product vendor for product images, etc.

Logically unwinding the reasoning behind all this irregular involvement and inefficiency in the workflow design only leads to frustration. What the workflow demands is repeatability.

Setting the Stage

Now consider a data governance kickoff meeting. We initiate the meeting by inviting participants and setting an agenda. We adjust the agenda to account for scheduling and the priorities of those invited. We need to plan carefully, because we don't want to remove a name from the invitation or run the risk of not inviting someone important and relevant. We build the invitation list carefully and with the involvement of the MDM project executive sponsor.

Generally, I invite representation from all subject areas, the currently scoped and to-be scoped business areas we are covering with our MDM program.

If we include every business area that rolls up to the stock ticker, we're probably creating far too broad a scope for data governance. Depending on the size of the organization, we might need multiple governance organizations that roll up in a hierarchical sort of way. Furthermore, existing management groups that have tangential function to data governance may need to be included. Don't ignore them, and keep in mind that you might be able to leverage an existing council, committee or steering group for data governance.

When setting expectations and plotting the course of progress for data governance, the culture of the organization must be taken into account. With data governance, you are naturally adding a measure of structure in an area where structure is desirable and needed. In highly collaborative environments where decision-making is drawn out (an unacceptable hindrance to data governance), you may be broaching the topic of decisioning protocol for the first time for the organization. This brings data governance to a new level of required rigor.

Readiness proceeds by designing short-term targets for data governance - decisions the MDM team needs now in order to develop workflow this quarter. Be sure that all decisions are not only logged, but also reflected in the workflow. Governance participants' enthusiasm will fade if they do not see the result of their work manifested in their actual daily activity.

While we're not likely to overestimate an organization's overall readiness for data governance, spending any time under false presumptions is wasteful, and wasted time always comes back to sting your efforts. It is possible that certain parts of the organization are more ready than others. Understand the readiness of your organization for data governance as the first step in implementation.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access